Hot Carrier Extraction with Plasmonic Broadband Absorbers - ACS

Mar 16, 2016 - Ultra-broadband perfect solar absorber by an ultra-thin refractory titanium nitride meta-surface. Zhengqi Liu , Guiqiang Liu , Zhenping...
0 downloads 6 Views 1MB Size
Subscriber access provided by NEW YORK UNIV

Article

Hot Carrier Extraction with Plasmonic Broadband Absorbers Charlene Ng, Jasper Cadusch, Svetlana Dligatch, Ann Roberts, Timothy J Davis, Paul Mulvaney, and Daniel E. Gomez ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.6b01108 • Publication Date (Web): 16 Mar 2016 Downloaded from http://pubs.acs.org on March 18, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Nano is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Hot Carrier Extraction with Plasmonic Broadband Absorbers Charlene Ng,∗,† Jasper Cadusch,¶ Svetlana Dligatch,§ Ann Roberts,¶ Timothy J. Davis,¶ Paul Mulvaney,k and Daniel E. G´omez∗,† †CSIRO, Manufacturing, Private Bag 33, Clayton, VIC, 3168, Australia ‡Melbourne Centre for Nanofabrication, Australian National Fabrication Facility, Clayton VIC 3168, Australia ¶School of Physics, The University of Melbourne, Parkville, VIC, 3010, Australia §CSIRO, Manufacturing, PO Box 218, Lindfield NSW 2070, Australia kBio21 Institute & School of Chemistry, The University of Melbourne, Parkville VIC 3010, Australia E-mail: [email protected]; [email protected]

Abstract Hot charge carrier extraction from metallic nanostructures is a very promising approach for applications in photo-catalysis, photovoltaics and photodetection. One limitation is that many metallic nanostructures support a single plasmon resonance thus restricting the light-to-charge-carrier activity to a spectral band. Here we demonstrate that a monolayer of plasmonic nanoparticles can be assembled on a multistack layered configuration to achieve broad-band, near-unit light absorption, which is spatially localised on the nanoparticle layer. We show that this enhanced light absorbance leads to ∼ 40–fold increases in the photon–to–electron conversion efficiency by the plasmonic nanostructures. We developed a model that successfully captures the essential physics

1

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of the plasmonic hot–electron charge generation and separation in these structures. This model also allowed us to establish that efficient hot carrier extraction is limited to spectral regions where: (a) the photons have energies higher than the Schottky junctions and (b) the absorption of light is localised on the metal nanoparticles. Keywords: Surface plasmons, hot–spots, surface enhancements, near fields, photocatalysis.

2

ACS Paragon Plus Environment

Page 2 of 25

Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Metal nanoparticles exhibit absorption cross-sections for incident light that can far exceed their physical dimensions. This strong interaction with light originates from the excitation of collective oscillations of surface charges in the nanoparticles, commonly referred to as surface plasmons. Plasmons can relax radiatively by re-emitting photons or non-radiatively creating a transient population of non-equilibrium (hot) chargecarriers, 1–5 which in a chemically inert environment, transfer their energy to the metal lattice resulting in nanoparticle heating. 6 However, it is also possible for these hot charge carriers to be deposited into acceptor states of adsorbates 7 or to be transferred to an accepting medium creating charge–separated states with sufficient chemical potential energy to drive chemical (redox) reactions. For instance, in metal-semiconductor Schottky junctions, 8 plasmon relaxation can result in the emission of hot charge carriers into the conduction band of the semiconductor, 9,10 a charge–separation process that requires photon energies below the band gap energy of the semiconductors and which has been used in photovoltaics, 8 photodetection 11 and photocatalysis. 10,12–14 Plasmonic hot charge–carrier generation has recently been used for the light-driven dissociation of hydrogen, 15,16 oxidation of ethylene, 17 cross–coupling reactions, 18 reduction of nitro–aromatics to azo compounds 19 and other reactions involving more complex organic molecules. 20–22 The plasmon hot–carrier relaxation pathway is a process that depends on: 23 (i) the efficiency of the charge transfer that must take place at relevant interfaces, (ii) the charge separation following light absorption and (iii) the absorption of incident photons. Therefore, it is expected that the rate of plasmonic hot–charge carrier relaxation can be substantially increased under conditions where the metal nanostructures absorb nearly all incident photons across a broad spectral bandwidth. 24,25 Typically, near unit broadband absorption has been achieved with metal-insulator-metal (MIM) structures. 26–29 These consist of a thin insulator or dielectric layer (< 100 nm) sandwiched between a metal mirror and a thin top layer of light absorbing metal nanostructures. 30,31 By replacing the insulator layer in an MIM structure with a semiconductor such as TiO2 , it should be possible to extract plasmonic hot–carriers using a Schottky junction that

3

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

exhibits near unit broadband absorption. 24 In addition to the expected increases in hot–carrier generation efficiencies, near–unit broadband absorption is an ideal model system to study the mechanism of photo–conversion in plasmonic structures. Here we present a study on the extraction of plasmonic hot charge carriers from a monolayer of Au nanoparticles capable of absorbing up to > 90% of incident visible light in a metal-semiconductor-metal (MSM) configuration. Examinations of the photoelectrochemical behaviour of such structures reveal significant enhancements in the yields of hot carrier extraction. 25 Furthermore, with the aid of a simple phenomenological model, we establish that plasmon non–radiative relaxation yields uniform energy and momentum distributions of the generated charge carriers, which limit the spectral bandwidth of hot carrier extraction. 8 Results and Discussion Figure 1a shows the metal–semiconductor–nanoparticle structures, which consist of an optically thick metal film (200 nm) and a single layer of Au nanoparticles, separated by a thin 50 nm TiO2 layer. These samples were made by means of sequential physical vapour deposition of materials (see Methods section), a large scale method resulting in a densely–packed monolayer of Au nanoparticles [shown in the scanning electron microscope (SEM) image of Figure 2a: 10 nm thick, average diameter of 20 nm, dimensions that are ≤ mean-free path for hot electrons in Au. 33 Size histogram in Supplementary Figure S1]. This nanoparticle ensemble, when deposited on top of glass–supported TiO2 films, exhibits a single–band absorption centred around 620 nm (with 0.15 absorbance) characteristic of localised surface plasmon resonances of Au nanoparticles on materials with a high refractive index. 13 On the contrary, the metal– semiconductor–nanoparticle structure exhibits a distinctively broadband absorption in a spectral region ranging from 600 nm to 1000 nm, reaching absorbance values of up to ∼0.75: a 5-fold increase in absorbance, when compared to the control sample without the reflecting layer (no metal). These absorption spectra were calculated as A = 1 − R − T , where R and T correspond to measured reflectance and transmission spectra (both their diffuse and specular components, see Methods section).

4

ACS Paragon Plus Environment

Page 4 of 25

Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Figure 1: Architecture and working principle of plasmonic broadband absorber structure. (A) Schematic of the broadband absorber metal-semiconductor-nanoparticle structure with TiO2 positioned in between the Au mirror and Au nanoparticles. Inset: multiple internal reflection and interference interpretation of unit absorption. 32 Also shown is an SEM of the Au nanoparticle monolayer. (B) Absorbance spectra of metal-semiconductornanoparticle, no nanoparticle, no metal and bare TiO2 . These spectra were calculated as A = 1 − R − T , where R and T are the measured Reflectance and Transmission spectra (diffuse + specular). (C) Spatial distribution of the Electric field of the metal–semiconductor– nanoparticle structures at 650 nm. The values are shown on a logarithmic scale of the absolute magnitude square of the electric field relative to the incident field. (D) Plot of the absorption of incident light vs. wavelength for the top and bottom metal layers.

5

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In the metal–semiconductor–nanoparticle structures, the thickness of the TiO2 layer was chosen, with the aid of numerical solutions of Maxwell equations (see supporting information section S1) such that the structure exhibited low reflection (R) of light in the visible. Given that the thickness of the supporting metal layer is much greater than the skin depth of the metal, there is no transmission of light in these structures (T = 0) and conservation of energy dictates that the absorbance A is therefore given by A = 1 - R, implying that (near) unit absorption occurs when R (tends to) equals zero. Near–unit absorption in these structures originates from optical destructive interference between directly reflected light and the multiple reflections taking place at the other interfaces of the multi–layer stack 32 (see diagram of Figure 1A). Other interpretations assign the near–unit absorption of light to a renormalization of the polarizability of the nanoparticles due to their interactions with the mirror 34 or to optical impedance matching between the structure and free space, a phenomenon that occurs due to a magnetic response that arises from anti-parallel currents taking place on both the metal mirror and the nanoparticles. 26–29 According to the numerical simulations shown in Figure 1C, at a wavelength of 650 nm there is a strong (and sub–wavelength) localisation of the electric field at the spatial location of the metal nanoparticles, which accounts for the high absorbance reported in Figure 1B. As a function of the incident photon wavelength, Figure 1D shows that the spatial location of the absorption occurs at the nanoparticle monolayer for long wavelengths, but it becomes localised on the metal mirror at shorter wavelengths. Subsequent to light absorption, non–radiative relaxation of nanoparticle plasmons can result in the transfer of the incident photon energy to a hot electron, which can obtain sufficient energy and momentum to overcome the Schottky barrier existent at the Au nanoparticle–TiO2 interface. 35 To measure this possible photon–to–electron conversion, we employed the metal–semiconductor–nanoparticle structures as photoanodes in photo–electrochemical cells, as illustrated in the diagram of Figure 2A. 36–38 Electron injection from the metal nanoparticles into accepting states (such as the conduction band

6

ACS Paragon Plus Environment

Page 6 of 25

Page 7 of 25

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

or trap states) of the TiO2 film, leaves a positive charge (hole) on the Au nanoparticles which can be neutralised by electron donating species in solution. Injected electrons in the TiO2 can then be collected by the mirror (which also acts as an electrical contact) and transported to a Pt counter electrode, where reduction reactions take place. The overall process results in the generation of measurable photocurrents. Figure 2B shows the measured photocurrents for three photoanodes: (i) metal– semiconductor–nanoparticle , (ii) metal–supported TiO2 thin film (no nanoparticles), and (iii) a semiconductor–supported Au nanoparticle monolayer (no metal). These measurements were performed under visible light irradiation (495 nm long pass filter) in a two-electrode configuration with an applied potential of 0.5 V, and under conditions where the electrolyte consisted of aqueous 0.5 M Na2 SO4 with 20 v/v% methanol (purged with N2 , more details can be found in the methods section). The long-pass filter eliminates possible direct excitation of electron–hole pairs in TiO2 with photon energies above its band gap (see Supplementary Figure S2 where we estimate the band-gap of the TiO2 ). Indeed, significant photocurrents at wavelengths higher than 495 nm were only observed when Au nanoparticles were present on the electrodes. The photocurrents observed in Figure 2B can only be a consequence of the excitation and extraction of hot charge–carriers from the Au nanoparticles into the Pt counter electrode. These photocurrents were detected only when the visible light was turned on and the currents returned to the background in the dark. This confirms that the measured electrical currents were photo–induced, as opposed to dark (thermal) currents enhanced by the applied bias voltage [see Figure S4 where the effect of the applied bias on the measured photocurrents is reported]. The measurements were reproducible and stable for all of the samples, indicating the absence of irreversible electrochemical damage to the photoanodes. The anodic currents measured with the metal–semiconductor–nanoparticle sample (0.56 µA cm−2 ) were 3.5 times larger than those obtained with the semiconductor–supported Au nanoparticle monolayer (no metal) (0.16 µA cm−2 ) electrode. The magnitude of these photocurrents increased linearly with the incident illumination power (see Supplementary Figures

8

ACS Paragon Plus Environment

Page 8 of 25

Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

S3). The IPCE as a function of wavelength was measured for all three samples, and the results are shown in Figure 2C. The metal–supported TiO2 thin film (no nanoparticle) did not lead to an appreciable IPCE even though it showed a relatively high absorption at 460 nm. On the contrary, the non-vanishing IPCE spectrum measured for the semiconductor–supported Au nanoparticle monolayer (no metal) closely follows the lineshape of its corresponding optical absorption spectrum shown in Figure 1B, indicating a strong correlation with the photoexcitation of localised surface plasmon resonances. For the metal–semiconductor–nanoparticle photo–anode, the measured IPCE has a broadband character with a maximum of 0.27 % at around 600 nm, an efficiency ∼ 20 times larger than the one measured for the semiconductor–supported Au nanoparticle monolayer (no metal). Further increases in the absorbance and (consequently) IPCE values of the metal– semiconductor–nanoparticle structures can be achieved by decreasing the thickness of the TiO2 layer from 50 nm to 30 nm, which as shown in Figure 3(A), leads to an absorbance of up to ≥ 95% (maximum of 99% at 544 nm). The IPCE in turn increases by about a factor of 4× for wavelengths below 600 nm, but remains almost unchanged at longer wavelengths. In Figure 3(B), we show that when the mirror is modified from Au to Al in the metal–semiconductor–nanoparticle stacks (at fixed semiconductor thickness), the measured light absorption increases to ≥ 80% (and reaches a maximum of 87% at 590 nm) with concomitant increases in IPCE values. Figure 3(C) depicts a summary of the IPCE enhancements attainable with the metal– semiconductor–nanoparticle structures investigated, which shows increases of almost 40× with respect to the semiconductor–supported Au nanoparticle monolayer. The measured IPCE spectral lineshapes for the metal–semiconductor–nanoparticle samples track the corresponding optical absorption spectra for wavelengths shorter than ∼ 700 nm, but exhibit a characteristic rapid decay at longer wavelengths (in spite of measured strong absorbance) which occurs as a consequence of the mechanism of hot–charge carrier extraction that we proceed to discuss next.

9

ACS Paragon Plus Environment

ACS Nano

(A) 100 30nm TiO2

10

50nm TiO2

1 10 -1 10 2 400

500

600

700

800

(B) 100 Al Au

%

10 1 10 -1 10 2 400

(C)

500

600

700

800

Wavelength (nm)

80

IPCE ratio

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 25

30nm TiO2

60

50nm TiO2

40

Al

20 0 400

500

600

700

Wavelength (nm)

Figure 3: Effect of structural parameters on measured IPCE. (A)TiO2 thickness. Dashed lines are the measured optical absorption spectra of samples (obtained from diffuse reflectance measurements) with 30 nm and 50 nm of TiO2 . Dotted lines show the corresponding IPCE values. (B) Work function of the reflector. Measured absorption spectra for samples where the metal reflector was made of Au or Al and their corresponding IPCE spectra (dotted lines). The thickness of TiO2 was kept fixed at 50 nm. (C) Ratio of measured IPCE relative to the results obtained for the TiO2 supported Au nanoparticle monolayer.

10

ACS Paragon Plus Environment

Page 11 of 25

(A) 10 0 P (Ee > φB ) P (Ee > φB |ke ) ⊗ hT i

Probability

10 -1

10 -2 ΦSB = 1.35eV

10

-3

10 -4 0.5

1

1.5

2

2.5

3

Photon energy (eV)

(B) 10 2 IPCEexp. η(Φ SB =1.20eV)

Probability (%)

10 1

η(Φ SB =0.90eV) η(Φ SB =1.35eV) A layer

10 0

10 -1

10 -2

1

1.5

2 Energy (eV)

2.5

(C) 10 0

IPCE (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

MeOH EtOH IPA

10 -1

10 -2 400

500

600

700

800

Wavelength (nm)

Figure 4: Estimate of the efficiency. (A) hot–electron injection probability ηinj = P (Ee > ΦSB |ke ) ⊗ hT i. P (Ee > ΦSB |ke ) is the joint probability that plasmonic hot–electrons have energies Ee in excess of the Schottky barrier ΦSB and a momentum ke with a component perpendicular to the metal–semiconductor interface that lies within the escape cone. hT i is the electron transmission coefficient at the nanoparticle–TiO2 interface. Also shown for comparison, is the probability P (Ee > ΦSB ) that the plasmon–derived hot–electrons have energies Ee in excess of ΦSB . (B) η: absorbed photon–to–electron conversion efficiency. It shown in this figure for two values of ΦSB . The solid curves represent the theoretical estimate of the efficiency given by Alayer × η, where Alayer (dashed grey line) is the calculated absorption at the nanoparticle layer. The modelled and measured IPCE (IPCEexp ) show similar lineshapes exhibiting a pronounced decrease as the wavelength of incident photons approaches ΦSB , consistent with the data shown in (A). In these calculations values of 0.96 eV, 39 1.20 eV 40 and 1.35 eV were used for ΦSB . (C) Measured IPCE spectra for cases with Methanol (MeOH), Ethanol (EtOH) and Isopropanol (IPA) as the sacrificial electron donors 11 in solution. ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

The IPCE at a specific wavelength λ is proportional to the photon absorption efficiency of the metal nanoparticles Alayer and the probability η that the absorbed photon results in a collected electron in the photoelectrochemical circuit: 33

IPCE(λ) = Alayer (λ) × η(λ).

(1)

The former can be determined by solving Maxwell’s equations for the structures, which also provide information on, for instance, the spectral and spatial location of the lightabsorption events (e.g. figure 1D). Alayer is optimal for structures that exhibit unit absorption of light at the position of the nanoparticle monolayer (i.e. Alayer = 1). In order to get a quantitative estimate for η, as a first–order approximation, this parameter can be expressed as (more details in Supplementary section S3):

η = ηinj × ηtrpt × ηinjm × ηed ,

(2)

which conceptually results from the following sequence of events (see the diagram shown in Figure 2): (i) the injection of a plasmon–derived hot electron across the nanoparticle–semiconductor Schottky barrier, with probability ηinj , (ii) elastic transport of the injected electrons across the semiconductor layer with an associated probability ηtrpt , (iii) the injection of the charge carrier into the metal reflector, occurring with a probability ηinjm . An additional assumption is made in that once injected into the reflector, the charge carrier will travel through an external circuit to the counter electrode where finally (iv) the total current is measured when an electron–donating species in solution injects an electron into the positively–charged metal nanoparticles with a probability ηed . 23 The hot–electron injection efficiency ηinj is determined by the energy–momentum distribution of the hot–electron population that results from Landau damping of plasmons. Little is known at present about these energy–momentum distributions. To get an estimate of ηinj , we assume that Landau damping produces an isotropic momentum distribution 41 and approximate the energy distribution of hot–electrons as a

12

ACS Paragon Plus Environment

Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

product of the initial and final (parabolic) density of electron states in the metal. We furthermore consider the initial states to have energies ranging from below (and up to) the Fermi level EF , whereas the final states are thought to have energies from the Fermi level and up to EF + the incident photon energy. 42 ηinj is then approximated as P (Ee > ΦSB |ke ) ⊗ hT i [see equation (S12)], where: hT i is the transmission coefficient for electrons across the metal nanoparticle–semiconductor interface (calculated taking into account conservation of momentum, see section S3.1.3) and, P (Ee > ΦSB |ke ) is the fraction of the hot–electron population for which the electrons have a momentum ke that lies within the escape cone of the metal–semiconductor interface and with energies Ee above the Schottky barrier [see equation (S7)]. As shown in Figure 3A, ηinj increases slowly with incident photon energy [details of the calculations leading to Figure 4(A) are shown in the supplementary section S3, note that we do not consider possible quantum mechanical tunnelling of electrons under the barrier]. The charge–carrier transport efficiency ηtrpt is the probability that, after injection, the charge carrier is transported through the TiO2 layer, without experiencing scattering and/or trapping at defect states, the probability of which we assume to be determined by the ratio of the thickness of the metal–oxide layer and the mean–free path for electrons 43 (see see Supplementary information section S3.2). ηinjm , is approximated by the transmission coefficient for electrons to traverse the semiconductor–metal (reflector) interface, thus accounting for possible reflections due to the momentum and energy mismatch (more details in in Supplementary Information section S3.3). For simplicity, we do not account for possible electron flow from the metal reflector into the semiconductor, which may originate from the excitation and decay of surface plasmon polaritons at the metal/semiconductor interface. This charge flow will have an opposite direction and thus decrease the measured photocurrents and IPCE. 44,45 The calculated values of η (assuming ηed = 1) and Alayer ×η produce the dotted and continuous lines in Figure 4B respectively. η is strongly dependent on the magnitude of ΦSB , the thickness of the oxide layer and the height of the energy barrier preventing electron injection at the semiconductor–mirror interface. The model satisfactorily

13

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

reproduces the slow increase of the measured IPCE with incident photon energy, in particular, for a value of ΦSB = 1.20 eV reported by the experiments of Lee et al. 46 In the model, this curvature originates from a combination of two factors: (i) the almost rectangular shape of the energy distribution of hot carriers, which limits the number of these carriers that meets the energy requirements for injection [i.e. E > ΦSB , see curve P (Ee > ΦSB ) of Figure 4A] and, (ii) the narrow escape cone for carriers from the metal nanoparticles and into the semiconductor, which severely limits the total fraction of “useful” charge carriers by almost 75% [see Figure S9(B)]. Factor (i) is a consequence of the assumed broad energy distribution of initial states, whereas (ii) originates from the assumed uniform distribution of electron momenta resulting from Landau damping. At higher energies, the calculated drop in Alayer × η is in qualitative agreement with the experimental results, which is a consequence of the fact that at these wavelengths, light is predominantly absorbed by the mirror (see Figure 1D). A simple conclusion that can be drawn from these observations is that, in spite of the spectrally broad and strong absorption of light achievable with the metal–semiconductor–nanoparticle structure, the generation of charge–carriers from plasmon relaxation in this architecture is limited to those spectral regions where the incident photon energies are in excess of the metal–semiconductor Schottky barrier and below the onset for strong absorption by the metal reflecting layer. According to the model, the ICPE is expected to increase if the plasmon relaxation were to result in narrow electron energy distributions. This is a situation that can be modelled by considering that the bottom of the conduction band is close (∼ 0.15 eV) to the Fermi level of the metal, 42,47 a hypothetical (and somewhat unrealistic) situation depicted in Figure 5. The orange curve in Figure 5(A) shows that for this hypothetical case, the fraction of the population of hot–electrons with Ee > ΦSB is almost a step function of the incident photon energy and becomes energy–independent (and unit) for energies slightly above ΦSB . This step–like P (Ee > ΦSB ) drastically changes the shape of η [Figure 5(B)] and leads to an overall increase in the expected internal efficiency The measured IPCEs are well below the predictions of our model, which were made

14

ACS Paragon Plus Environment

Page 14 of 25

Page 15 of 25

(A) 1 E F = 5.1 eV

e

P(E >Φ

SB

)

0.15 eV

0.5

0 0.5

1

1.5

2

2.5

Photon energy (eV) 10

η (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

1

(B)

10 0 10 -1 10 -2 0.5

1

1.5

2

2.5

Photon energy (eV) Figure 5: Narrow distribution of states near the Fermi level of the metal. (A) Distribution of electron energies and (B) calculated internal quantum efficiency. The shaded areas highlight the attainable differences. ΦSB = 0.98 eV.

15

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

under the assumption of unit ηed . ηed in the photo–electrochemical configuration that we adopted for extracting hot–charge carriers is expected to be limited by the kinetics of charge and mass transfer occurring at the solid–liquid interfaces. In Figure 4C we show the measured IPCE changes for a sample where the (sacrificial) electron donating species were changed from Methanol (MeOH) to Ethanol (EtOH) and to Isopropanol (IPA). An uniform increase in the IPCE is clearly observable for the case of EtOH, which is attributable solely to changes in the kinetic and thermodynamic processes taking place at the photoanode/liquid junction. ηed is a parameter which is not included in our theoretical description of plasmonic hot–charge carrier extraction, but which seems to play a key role in determining the IPCE in our photo–electrochemical cell, an observation also conjectured in (solid state) plasmonic solar cells, 48 where it was found that hole accumulation at the metal nanoparticles limits device performance. Conclusions In summary, we have demonstrated the extraction of plasmonic–derived charge carriers from a multi–layer stack comprising a monolayer of Au nanoparticles. These multilayer structures exhibit broadband and intense absorption of light, which leads to a significant increase in the incident photon–to–electron conversion efficiencies. We have developed a simple model to describe the hot–charge carrier generation and transport in this system, which satisfactorily describes the measured IPCE spectra. According to this model, the broad distribution of hot–charge carrier energies and their uniform distribution of momenta, account for the weak increase in measured IPCE with incident photon energies. A more efficient strategy for generating and extracting plasmonic hot charge carriers is one that combines strong near–unit absorption of light with a mechanism, such as the plasmon-induced metal-to-semiconductor interfacial charge transfer transition, where a quantum yield for electron injection of >24% has been found to be independent of the incident photon energy. 49 Our results demonstrate that significant enhancements in the efficiencies of opto–electronic and photoelectrochemical devices that operate with plasmonic–derived hot–charge carriers can be achieved by tailoring the optical properties of the plasmonic nanostructures.

16

ACS Paragon Plus Environment

Page 16 of 25

Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Methods/Experimental Material fabrication and Characterization The Al and Au mirrors were deposited onto glass substrates with an in–house constructed electron beam evaporator equipped with transmittance monitoring. A thin layer of Cr was deposited prior to the Au deposition to ensure good adhesion to the glass substrate. For both Al and Au mirrors, deposition was conducted by monitoring the optical transmittance to ensure the films had no transmittance of light. Subsequently, the TiO2 layers (30 nm and 50 nm) were deposited onto the metal supporting mirrors through an ion–assisted electron beam evaporation process at 200◦ C. The bombardment of oxygen during the deposition process ensured the deposition of stoichiometric TiO2 onto the mirrors. Lastly, the Au nanoparticles were deposited with a similar electron beam evaporation process with two types of in–situ optical monitoring. A broadband transmittance was utilized to monitor the absorption of the Au nanoparticles, while an ellipsometry measurement at 633 nm was employed to achieve non–overlapping particles and detect the deposition point at which the particles coagulate to form continuous films. Scanning electron microscope (SEM) images of the Au NPs were obtained with a Zeiss Merlin Field Emission Scanning Electron Microscope. The diffuse and specular reflectance (R) and transmittance (T ) spectra were measured using a UV-VIS spectrophotometer (Perkin-Elmer Lambda 1050) with an integrating sphere and small spot kit. With these two measurements, the absorbance was calculated as A = 1 − R − T . Photoelectrochemical measurements The designed structure and platinum wire was employed as the working electrode and counter electrode respectively in a 2 electrode system. The exposed surface area of the working electrode was 1 cm in diameter. The electrolyte solution used was 0.5 M Na2 SO4 (anhydrous, Sigma-Aldrich, ≥99%) with 20 v/v % amount of methanol (Sigma-Aldrich, ≥99.9%) or ethanol (Sigma-Aldrich, ≥99.9%) as the sacrificial reagent. The 0.5 M Na2 SO4 was employed to reduce the resistance in the electrolyte solution. Prior to the photocurrent measurements, the solutions were purged with N2 gas to remove electron scavenging O2 . The working potential was set at +0.5 V versus the Pt wire and the devices were illuminated with light from a 300 W Xenon lamp (Newport Model no. 669092) using a ≥ 495 nm cut-off filter (Thorlabs FGL-495). The applied voltage and photocurrent I were recorded with a potentiostat (AutoLab PGSTAT204). To measure the IPCE values, light from the Xenon lamp was coupled to a monochromator with a bandwidth of 5 nm at full width at half maximum (FWHM). The light intensity P of the monochromatic light at each wavelength was measured using a Thorlabs optical power and energy meter (Model. PM100D).

17

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The incident photon to electron efficiency (IPCE) was calculated by the following formula:

IPCE = 100 ×

I(A/cm2 ) 1240 × . P (W/cm2 ) λ(nm)

18

ACS Paragon Plus Environment

Page 18 of 25

Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

References 1. Clavero, C. Plasmon-Induced Hot-Electron Generation At Nanoparticle/MetalOxide Interfaces For Photovoltaic And Photocatalytic Devices. Nat. Photonics 2014, 8, 95–103. 2. Brongersma, M. L.; Halas, N. J.; Nordlander, P. Plasmon-induced Hot Carrier Science And Technology. Nat. Nanotechnol. 2015, 10, 25–34. 3. Linic, S.; Christopher, P.; Ingram, D. B. Plasmonic-Metal Nanostructures For Efficient Conversion Of Solar To Chemical Energy. Nat. Mater. 2011, 10, 911– 921. 4. Manjavacas, A.; Liu, J. G.; Kulkarni, V.; Nordlander, P. Plasmon-Induced Hot Carriers in Metallic Nanoparticles. ACS Nano 2014, 8, 7630–7638. 5. Zhang, W.; Govorov, A. O.; Bryant, G. W. Semiconductor-Metal Nanoparticle Molecules: Hybrid Excitons and the Nonlinear Fano Effect. Phys. Rev. Lett. 2006, 97, 146804. 6. Hartland, G. V. Optical Studies of Dynamics in Noble Metal Nanostructures. Chem. Rev. 2011, 111, 3858–3887. 7. Kale, M. J.; Avanesian, T.; Christopher, P. Direct Photocatalysis by Plasmonic Nanostructures. ACS Catal. 2014, 4, 116–128. 8. Mubeen, S.; Lee, J.; Lee, W.-r.; Singh, N.; Stucky, G. D.; Moskovits, M. On the Plasmonic Photovoltaic. ACS Nano 2014, 8, 6066–6073, PMID: 24861280. 9. Furube, A.; Du, L.; Hara, K.; Katoh, R.; Tachiya, M. Ultrafast Plasmon-Induced Electron Transfer from Gold Nanodots into TiO2 Nanoparticles. J. Am. Chem. Soc. 2007, 129, 14852–14853. 10. Kim, H. J.; Lee, S. H.; Upadhye, A. A.; Ro, I.; Tejedor-Tejedor, M. I.; Anderson, M. A.; Kim, W. B.; Huber, G. W. Plasmon-Enhanced Photoelectrochemical

19

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Water Splitting with Size-Controllable Gold Nanodot Arrays. ACS Nano 2014, 8, 10756–10765, PMID: 25268767. 11. Knight, M. W.; Sobhani, H.; Nordlander, P.; Halas, N. J. Photodetection with Active Optical Antennas. Science 2011, 332, 702–704. 12. Nishijima, Y.; Ueno, K.; Kotake, Y.; Murakoshi, K.; Inoue, H.; Misawa, H. NearInfrared Plasmon-Assisted Water Oxidation. J. Phys. Chem. Lett. 2012, 3, 1248– 1252. 13. Tian, Y.; Tatsuma, T. Mechanisms and Applications of Plasmon-Induced Charge Separation at TiO2 Films Loaded with Gold Nanoparticles. J. Am. Chem. Soc. 2005, 127, 7632–7637. 14. Qian, K.; Sweeny, B. C.; Johnston-Peck, A. C.; Niu, W.; Graham, J. O.; DuChene, J. S.; Qiu, J.; Wang, Y.-C.; Engelhard, M. H.; Su, D. et al. Surface PlasmonDriven Water Reduction: Gold Nanoparticle Size Matters. J. Am. Chem. Soc. 2014, 136, 9842–9845, PMID: 24972055. 15. Mukherjee, S.; Libisch, F.; Large, N.; Neumann, O.; Brown, L. V.; Cheng, J.; Lassiter, J. B.; Carter, E. A.; Nordlander, P.; Halas, N. J. Hot Electrons Do the Impossible: Plasmon-Induced Dissociation of H2 on Au. Nano Lett. 2013, 13, 240–247. 16. Mukherjee, S.; Zhou, L.; Goodman, A. M.; Large, N.; Ayala-Orozco, C.; Zhang, Y.; Nordlander, P.; Halas, N. J. Hot-Electron-Induced Dissociation of H2 on Gold Nanoparticles Supported on SiO2 . J. Am. Chem. Soc. 2014, 136, 64–67. 17. Christopher, P.; Xin, H.; Linic, S. Visible-Light-Enhanced Catalytic Oxidation Reactions On Plasmonic Silver Nanostructures. Nat. Chem. 2011, 3, 467–472. 18. Xiao, Q.; Sarina, S.; Bo, A.; Jia, J.; Liu, H.; Arnold, D. P.; Huang, Y.; Wu, H.; Zhu, H. Visible Light-Driven Cross-Coupling Reactions at Lower Temperatures

20

ACS Paragon Plus Environment

Page 20 of 25

Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

Using a Photocatalyst of Palladium and Gold Alloy Nanoparticles. ACS Catal. 2014, 4, 1725–1734. 19. Ke, X.; Zhang, X.; Zhao, J.; Sarina, S.; Barry, J.; Zhu, H. Selective Reductions Using Visible Light Photocatalysts Of Supported Gold Nanoparticles. Green Chem. 2013, 15, 236–244. 20. Wang, C.; Astruc, D. Nanogold Plasmonic Photocatalysis For Organic Synthesis And Clean Energy Conversion. Chem. Soc. Rev. 2014, 43, 7188–7216. 21. Scaiano, J. C.; Stamplecoskie, K. Can Surface Plasmon Fields Provide a New Way to Photosensitize Organic Photoreactions? From Designer Nanoparticles to Custom Applications. J. Phys. Chem. Lett. 2013, 4, 1177–1187. 22. Xiao, M.; Jiang, R.; Wang, F.; Fang, C.; Wang, J.; Yu, J. C. Plasmon-Enhanced Chemical Reactions. J. Mater. Chem. A 2013, 1, 5790–5805. 23. Moskovits, M. The Case For Plasmon-Derived Hot Carrier Devices. Nat. Nanotechnol. 2015, 10, 6–8. 24. Fang, Y.; Jiao, Y.; Xiong, K.; Ogier, R.; Yang, Z.; Gao, S.; Dahlin, A.; K¨all, M. Plasmon Enhanced Internal Photoemission In Antenna-Spacer-Mirror Based Au/TiO2 Nanostructures. Nano Lett. 2015, 15, 4059. 25. Robatjazi, H.; Bahauddin, S. M.; Doiron, C.; Thomann, I. Direct Plasmon-Driven Photoelectrocatalysis. Nano Lett. 2015, 26. Liu, Z.; Liu, X.; Huang, S.; Pan, P.; Chen, J.; Liu, G.; Gu, G. Automatically Acquired Broadband Plasmonic-Metamaterial Black Absorber during the Metallic Film-Formation. ACS Appl. Mater. Interfaces 2015, 7, 4962–4968, PMID: 25679790. 27. Aydin, K.; Ferry, V. E.; Briggs, R. M.; Atwater, H. A. Broadband PolarizationIndependent Resonant Light Absorption Using Ultrathin Plasmonic Super Absorbers. Nat. Commun. 2011, 2, 517.

21

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 25

28. Hedayati, M. K.; Javaherirahim, M.; Mozooni, B.; Abdelaziz, R.; Tavassolizadeh, A.; Chakravadhanula, V. S. K.; Zaporojtchenko, V.; Strunkus, T.; Faupel, F.; Elbahri, M. Design of a Perfect Black Absorber at Visible Frequencies Using Plasmonic Metamaterials. Adv. Mater. 2011, 23, 5410–5414. 29. Landy, N. I.; Sajuyigbe, S.; Mock, J. J.; Smith, D. R.; Padilla, W. J. Perfect Metamaterial Absorber. Phys. Rev. Lett. 2008, 100, 207402. 30. H¨ agglund,

C.;

Zeltzer,

G.;

Ruiz,

R.;

Thomann,

I.;

Lee,

H.-B.-R.;

Brongersma, M. L.; Bent, S. F. Self-Assembly Based Plasmonic Arrays Tuned by Atomic Layer Deposition for Extreme Visible Light Absorption. Nano Lett. 2013, 13, 3352–3357, PMID: 23805835. 31. H¨ agglund, C.; Apell, S. P. Plasmonic Near-Field Absorbers for Ultrathin Solar Cells. J. Phys. Chem. Lett. 2012, 3, 1275–1285, PMID: 26286771. 32. Chen, H.-T. Interference Theory Of Metamaterial Perfect Absorbers. Opt. Express 2012, 20, 7165–7172. 33. McFarland, E. W.; Tang, J. A Photovoltaic Device Structure Based On Internal Electron Emission. Nature 2003, 421, 616–618. 34. Kwadrin, A.; Osorio-A., C. I.; Koenderink, F. Backaction In Metasurface Etalons. Phys Rev B 2016, 93, 104301. 35. Mubeen, S.; Hernandez-Sosa, G.; Moses, D.; Lee, J.; Moskovits, M. Plasmonic Photosensitization of a Wide Band Gap Semiconductor: Converting Plasmons to Charge Carriers. Nano Lett. 2011, 11, 5548–5552. 36. Thomann, I.; Pinaud, B. A.; Chen, Z.; Clemens, B. M.; Jaramillo, T. F.; Brongersma, M. L. Plasmon Enhanced Solar-to-Fuel Energy Conversion. Nano Lett. 2011, 11, 3440–3446. 37. Tian, Y.; Tatsuma, T. Plasmon-Induced Photoelectrochemistry At Metal Nanoparticles Supported On Nanoporous Tio2 . Chem. Commun. 2004, 1810–1811.

22

ACS Paragon Plus Environment

Page 23 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Nano

38. Garc´ıa de Arquer, F. P.; Mihi, A.; Kufer, D.; Konstantatos, G. Photoelectric Energy Conversion of Plasmon-Generated Hot Carriers in Metal–Insulator– Semiconductor Structures. ACS Nano 2013, 7, 3581–3588, PMID: 23495769. 39. Zhang, X.; Chen, Y. L.; Liu, R.-S.; Tsai, D. P. Plasmonic Photocatalysis. Rep. Prog. Phys 2013, 76, 046401. 40. Lee, H.; Keun Lee, Y.; Nghia Van, T.; Young Park, J. Nanoscale Schottky Behavior Of Au Islands On Tio2 Probed With Conductive Atomic Force Microscopy. Appl. Phys. Lett. 2013, 103, 173103. 41. Scales, C.; Berini, P. Thin-Film Schottky Barrier Photodetector Models. IEEE J. Quant. Electron. 2010, 46, 633–643. 42. White, T. P.; Catchpole, K. R. Plasmon-Enhanced Internal Photoemission For Photovoltaics: Theoretical Efficiency Limits. Appl. Phys. Lett. 2012, 101, 073905. 43. Seah, M. P.; Dench, W. A. Quantitative Electron Spectroscopy Of Surfaces: A Standard Data Base For Electron Inelastic Mean Free Paths In Solids. Surf. Interface Anal. 1979, 1, 2–11. 44. Chalabi, H.; Schoen, D.; Brongersma, M. L. Hot-Electron Photodetection with a Plasmonic Nanostripe Antenna. Nano Lett. 2014, 14, 1374 – 1380. 45. Wang, F.; Melosh, N. A. Plasmonic Energy Collection through Hot Carrier Extraction. Nano Lett. 2011, 11, 5426–5430. 46. Lee, Y. K.; Jung, C. H.; Park, J.; Seo, H.; Somorjai, G. A.; Park, J. Y. Surface Plasmon-Driven Hot Electron Flow Probed with Metal-Semiconductor Nanodiodes. Nano Lett. 2011, 11, 4251–4255. 47. Chan, E.; Card, H.; Teich, M. C. Internal Photoemission Mechanisms At Interfaces Between Germanium And Thin Metal Films. IEEE J. Quant. Electron. 1980, 16, 373–381.

23

ACS Paragon Plus Environment

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

48. Reineck, P.; Lee, G. P.; Brick, D.; Karg, M.; Mulvaney, P.; Bach, U. A Solid-State Plasmonic Solar Cell via Metal Nanoparticle Self-Assembly. Adv. Mater. 2012, 24, 4750–4755. 49. Wu, K.; Chen, J.; McBride, J. R.; Lian, T. Efficient Hot-Electron Transfer By A Plasmon-Induced Interfacial Charge-Transfer Transition. Science 2015, 349, 632– 635. Acknowledgments This work was performed in part at the Melbourne Centre for Nanofabrication (MCN) in the Victorian Node of the Australian National Fabrication Facility (ANFF). C. N. was supported by an OCE Fellowship from CSIRO. D.E.G. acknowledges the ARC for support through a Future Fellowship (FT140100514) D.E.G. and T.J.D. acknowledge the ANFF for the MCN Technology Fellowships. Author contributions C.N. and S.D. prepared the samples. C.N. performed all the experiments. J.C. and A.R. performed the numerical simulations. All the authors contributed to writing the manuscript. Additional information: Supplementary Information Available: The material is available free of charge via the internet http://pubs.acs.org. Competing financial interests: The authors declare no competing financial interests.

24

ACS Paragon Plus Environment

Page 24 of 25

Page 25 of 25

ACS Nano

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment